PROJECT SUMMARY Background and significance: Features of the built environment and the availability of specific community resources constrain individuals? choices and thus may contribute to the adoption and maintenance of health- promoting behaviors, such as eating healthy diets and engaging in physical activity, that in turn may affect downstream biological (e.g., BMI) and clinical outcomes (e.g., cardiovascular disease). New technologies have rapidly increased the ability to link databases with geo-referenced information on environmental features to individual-level health data, thereby exponentially propelling research that examines associations between the built environment and health. This linked information is highly relevant to clinical practice, making it feasible to tailor interventions and treatments; and to novel study designs that use large databases, such as electronic health records, to investigate the joint effects of built environment factors and individual susceptibility on a range of health and health care outcomes. These linked data are also important for building evidence to support emerging urban design strategies that seek to create environments that promote health. However, methodological challenges, including exposure assessment and selection biases, make it difficult to identify the true impact of the built environment on individual behaviors, and the consequent ability to design place-based interventions to improve health. Objectives, innovation: This project uses data from a state of the art longitudinal study and methods that address residential self-selection, and develops and applies innovative approaches to address measurement challenges. The project will: (1a) estimate the geographic and temporal scales that are relevant for the effects of time-varying availability of community resources and built environment features on repeated measures of health behaviors, biological and clinical outcomes, and (1b) quantify differences in geographic and temporal scales across individuals and specific mechanisms; (2) develop and apply novel methods to ascertain and quantify exposures to multiple, complex environmental features, and assess their simultaneous impact on health indicators; (3) quantify the impact of measurement error in existing large scale built environment databases on estimated associations. The proposed analytical tools will be made freely available through R software packages, and results will be disseminated to multiple audiences, including urban planners. Impact: Results from this project will (a) increase precision in the translation, interpretation and evidence synthesis of past, current, and future studies of built environment health effects; (b) provide scientists with specific guidance on how to standardize measures of the built environment, and incorporate them into large scale individual-level health databases (e.g., electronic health records); and (c) inform best practices for the next generation of research on the impact of the built environment on population health. The project will result in approaches to more accurately and comprehensively measure the impact of multiple environmental factors on health.